>I have been giving some thought to how to include Ambisonics in Ogg Vorbis.As I understand it, mapping type = 1 was meant from the start to indicate an Ambisonic stream. The only other information required is the number of channels which thanks to Mr Leese's clever trick tells us exactly which Ambisonic channels are used up to 3rd order. __________________>The channel coupling wont be optimal and the phase may get a bit munged (Ambisonics is big on low-frequency phase), but it will work.Can we stop perpetuating this myth.>From the previous discussions on coupling and phase, it should be obvious (to yus DSP gurus) that all coupling methods that are likely to be used with an Ambisonic (or indeed any multichannel) stream will preserve phase.The phase mucking methods have not been implemented. __________________ So from the above, a Vorbis codec is straightforward. It only remains for a DSP guru to implement this for a common multi-channel player like VideoLAN or Windoz Media Player. Sebastien Olter is doing some code which will be useful towards this end. Windoz Media Player already has a good Ambisonic B-format to speaker decoder courtesy Bruce Wiggins' WAD http://sparg.derby.ac.uk/SPARG/Staff_BW.asp These should work with any DirectShow capable players too. If you are a DSP guru planning to do this for Windoz Media Player, please contact Bruce to make sure your decoder will feed WAD correctly in WMP. -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.446 / Virus Database: 269.4.0/762 - Release Date: 15/04/07 16:22
People, Depending on how many possible Ambisonic formats, variants, mappings do we have, things go more or less complicated. Why not just say: we encode B-Format and switch to mapping 1? Regarding the backward compatibility: Most of good Vorbis players and player libraries that can do multichannel decoding (for example: Winamp with 5.1, ffdshow, mplayer, vlc) are updated so often that it's possible to release a new valuable standard and simply wait for projects to follow. The only thing to be considered is the compatibility of already existing hardware players, those that have Ogg Vorbis 1 support built-in and disallow users to upgrade firmware. How many of them are out there? And who uploads a 300 kbps audio file (I assume Vorbis Ambisonic WXY with q=4) to a portable player? I think before the low bitrate Vorbis Ambisonic files become popular, such hardware players will all be out of service. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.xiph.org/pipermail/vorbis-dev/attachments/20070416/6e1b16da/attachment.html
On 4/16/07, Sebastian Olter <qduaty@gmail.com> wrote:> The only thing to be considered is the compatibility of already existing > hardware players, those that have Ogg Vorbis 1 support built-in and disallow > users to upgrade firmware. How many of them are out there? And who uploads a > 300 kbps audio file (I assume Vorbis Ambisonic WXY with q=4) to a portable > player? I think before the low bitrate Vorbis Ambisonic files become > popular, such hardware players will all be out of service.Fine enough. It would be nice if someone with a working ambisonic playback rig would give me some feedback on the decodes of coupled encodes. :)
On 16/04/07, Sebastian Olter <qduaty@gmail.com> wrote:> People, > > Depending on how many possible Ambisonic formats, variants, mappings do we > have, things go more or less complicated. Why not just say: we encode > B-Format and switch to mapping 1? > > Regarding the backward compatibility: ><snip the rest> I think, in general, this is probably the right approach to compatibility issues. What would be a useful goal would be a way to strip streams down to Vorbis I compatibility. -- imalone